Discovery of solid-state electrolytes for Na-ion batteries using machine learning

A machine learning predictive model based on Random Forest (RF) algorithm was built using experimental works reported on Na-ion solid electrolytes to discover new potential solid-state electrolytes with high ionic conductivity. The model was used to predict ∼25 K compounds from open materials databases and led to the identification of 4 compounds (NaPb3, Na3BiO3, Na2MoO4, NaMoF6) which were expected to show high ionic conductivity and supported by DFT calculations.